New approaches are needed to develop an effective vaccine for the prevention of HIV infection. Among the major challenges faced in designing an effective vaccine is the genetic variability of the virus and its ability to rapidly escape immune responses once it has disseminated within the infected host. HIV is most frequently acquired via mucosal exposure through sexual contact. Studies in macaques and humans have shown that SIV or HIV initially replicates locally in mucosal tissues prior to disseminating to lymphoid tissues throughout the body through the bloodstream. A vaccine that acts during early events following exposure at the local mucosal site may be more effective in limiting or preventing virus replication and subsequent dissemination. This application will take an approach to eliciting local mucosal immune responses through vaccination with parainfluenza virus type 5 (PIV5) vectors expressing SIV Gag and HIV Env and through mucosal application of Gag-Env virus-like particles (VLPs) engineered to reach the rectal mucosa-associated lymphoid tissue. PIV5 has not previously been evaluated as an HIV vaccine candidate, and is attractive due to its mucosal application and proven ability to protect at mucosal sites for other pathogens. Experiments in this proposal will define the ideal route to elicit potent intrarectal immune responses in mice and macaques. A low-dose intrarectal challenge experiment will be performed in macaques following mucosal immunization with PIV5 and rectal or intramuscular boosting with VLPs. These experiments will determine if vectors designed to stimulate potent local response in the rectal mucosa are protective from SHIV infection, and will seek to define correlates of mucosal immune protection.